The traffic sign image recognition system is one of the essential elements required by an autonomous vehicle, where these signs also serve as a guide for the autonomous car. Images of traffic signs will be captured by cameras on uncrewed vehicles. These images will then be processed for recognition, where the features in the traffic sign images will be extracted to obtain a characteristic for each traffic sign. Some of the obstacles that exist are the condition of traffic signs, such as changing color due to exposure to heat and rain, variations in lighting on traffic signs that occur at night and during the day, and also the various shapes of these traffic signs that must be considered when carrying out extraction. The features are round, rhombus, and square traffic signs. In this research, a method for extracting traffic sign image features has been developed using HSV and Wavelet Gabor. HSV obtains a sign image (region of interest) from the existing image. In contrast, Wavelet Gabor is used to recognize the type of sign detected. This research used a database of 20 kinds of signs with round, rhombus, and square shapes. The traffic sign image recognition system using Gabor wavelet as a feature extractor achieved an accuracy of 93.33% and 100% accuracy at night time.
Enhancing Traffic Signs Recognition Systems Through Gabor Feature Extraction Techniques
2024-11-22
643809 byte
Conference paper
Electronic Resource
English
Gabor Filter Approach to Joint Feature Extraction and Target Recognition
Online Contents | 2009
|Wiley | 2022
|A Gabor Feature Classifier for Face Recognition
British Library Conference Proceedings | 2001
|A Gabor feature classifier for face recognition
IEEE | 2001
|